Diastereoselective oxidative addition of dihydrogen to IrI(CO)((R)-BINAP) and [Ir(CO)2((R)-BINAP)][SbF6]

Article abstract of DOI:10.1021/ic061471a

Two chiral iridium(I) (R)-BINAP complexes, IrI(CO)((R)-BINAP) (1) and [Ir(CO)₂((R)-BINAP)][SbF₆] (2), have been synthesized and characterized, and their reactivity with dihydrogen has been studied. Complex 1 is formed on the addition of (R)-BINAP to [Bu₄N][IrI ₂(CO)₂] in toluene, and 2 is generated by the addition of AgSbF₆ to a solution of 1 in dichloromethane under CO. A structure determination of complex 2 confirms a square planar coordination geometry, while that of 1 reveals a significant tetrahedral distortion from the expected planar coordination. Additionally, the structure of 1 shows a disorder between iodide and CO ligands. The reaction of 1 with H₂ proceeds under kinetic control and shows a high degree of kinetic and thermodynamic selectivity; the kinetic product is formed by H2 addition across the P-Ir-CO axis of IrI(CO)((R)-BINAP) and yields two diastereomers which then convert over time to two more stable diastereomers which correspond to oxidative addition across the P-Ir-I axis. The kinetically favored diastereomers are formed in an initial ratio of 8.6:1, corresponding to a ΔΔ* of 1.27 kcal/mol. The reaction of H₂ with the C₂-symmetric complex 2 also leads to the formation of two diastereomers, with one favored over the other kinetically by a 9.9:1 ratio on extrapolation to t = 0. When these reactions are followed using parahydrogen NMR methods, only one of the initially formed diastereomers in each case is found to exhibit substantial parahydrogen-induced polarization in the hydride resonances at room temperature.

Full text of DOI:10.1021/ic061471a

Inorg. Chem. 2007, 46, 1196−1204
Diastereoselective Oxidative Addition of Dihydrogen to
IrI(CO)((R)-BINAP) and [Ir(CO)2((R)-BINAP)][SbF6]
Abdurrahman C
Richard Eisenberg*,†
¸
. Atesin,† Simon B. Duckett,‡ Christine Flaschenriem,† William W. Brennessel,† and
Department of Chemistry, UniVersity of Rochester, Rochester, New York 14627, and Department
of Chemistry, UniVersity of York, York YO10 5DD, U.K.
Received August 4, 2006
Two chiral iridium(I) (R)-BINAP complexes, IrI(CO)((R)-BINAP) (1) and [Ir(CO)2((R)-BINAP)][SbF6] (2), have been
synthesized and characterized, and their reactivity with dihydrogen has been studied. Complex 1 is formed on the
addition of (R)-BINAP to [Bu4N][IrI2(CO)2] in toluene, and 2 is generated by the addition of AgSbF6 to a solution of
1 in dichloromethane under CO. A structure determination of complex 2 confirms a square planar coordination
geometry, while that of 1 reveals a significant tetrahedral distortion from the expected planar coordination. Additionally,
the structure of 1 shows a disorder between iodide and CO ligands. The reaction of 1 with H2 proceeds under
kinetic control and shows a high degree of kinetic and thermodynamic selectivity; the kinetic product is formed by
H2 addition across the P
−Ir−CO axis of IrI(CO)((R)-BINAP) and yields two diastereomers which then convert over
time to two more stable diastereomers which correspond to oxidative addition across the P
−
Ir I axis. The kinetically
−
favored diastereomers are formed in an initial ratio of 8.6:1, corresponding to a ∆∆G* of 1.27 kcal/mol. The
reaction of H2 with the C2-symmetric complex 2 also leads to the formation of two diastereomers, with one favored
over the other kinetically by a 9.9:1 ratio on extrapolation to t
) 0. When these reactions are followed using
parahydrogen NMR methods, only one of the initially formed diastereomers in each case is found to exhibit substantial
parahydrogen-induced polarization in the hydride resonances at room temperature.
Introduction
dihydride products with the hydride ligands in mutually cis
positions. Previous work using IrX(CO)(dppe) (X ) Cl, Br,
I; dppe ) bis(diphenylphosphino)ethane) has shown that the
reaction proceeds under kinetic control, as shown in Scheme
1, leading to a kinetic dihydride product with >99%
selectivity followed by isomerization to the thermodynamic
dihydride in >90% yield.9 (This isomerization involves at
least two paths, one being a simple reductive elimination/
oxidative re-addition sequence and the other being a bimo-
lecular dihydride transfer10.) The basis of the stereoselectivity
was rationalized by the preferential addition to the Ir(I)
complex of H2 with its molecular axis parallel to the P-Ir-
CO axis of the complex (see Scheme 1) relative to addition
with H2 parallel to the P-Ir-X axis that would lead to the
thermodynamic isomer. This preference is based on elec-
tronic factors, most notably the ability of CO to stabilize
the “Ir(H2) trigonal bipyramidal” transition state of the H2
The activation of dihydrogen plays an important role in
transition metal catalyzed reactions such as homogeneous
hydrogenation and hydroformylation.1-8 For d8 Rh(I) and
Ir(I) catalysts and model systems, the activation of H2
proceeds by concerted oxidative addition to yield metal
* To whom correspondence should be addressed. E-mail: eisenberg@
chem.rochester.edu.
† University of Rochester.
‡ University of York.
(1) Heinekey, D. M.; Oldham, J. W. J. Chem. ReV. 1993, 93, 913-926.
(2) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University
Science Books: Mill Valley, CA, 1987.
(3) Jessop, P. G.; Morris, R. H. Coord. Chem. ReV. 1992, 121, 155-284.
(4) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis. The Applications
and Chemistry of Catalysis by Soluble Transition Metal Complexes,
2nd ed.; John Wiley and Sons, Inc.: New York, 1992.
(5) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th ed.;
John Wiley and Sons, Inc.: New York, 1988.
(6) Ungvary, F. Coord. Chem. ReV. 2003, 241, 295-312.
(7) Chaloner, P. A.; Esteruelas, M. A.; Joo, F.; Oro, L. A. Homogeneous
Hydrogenation; Kluwer Academic Publishers: Dordrecht, The Neth-
erlands, 1994.
(9) Johnson, C. E.; Eisenberg, R. J. Am. Chem. Soc. 1985, 107, 3148-
3160.
(10) Kunin, A. J.; Johnson, C. E.; Maguire, J. A.; Jones, W. D.; Eisenberg,
R. J. Am. Chem. Soc. 1987, 109, 2963-2968.
(8) Knowles, W. S. Acc. Chem. Res. 1983, 16, 106-112.
1196 Inorganic Chemistry, Vol. 46, No. 4, 2007
10.1021/ic061471a CCC: $37.00
© 2007 American Chemical Society
Published on Web 01/24/2007

DiastereoselectiWe OxidatiWe Addition of Dihydrogen
Scheme 1
at the metal center. Several such systems have been examined
in this context to see if the diastereoselectivity found for H2
addition is consistent with the impressive ee’s obtained in
asymmetric hydrogenations using the cationic Rh catalysts.19-22
In one study, Kunin et al. employed the chiraphos analogue
of the Ir(I) complex shown in Scheme 1 (chiraphos )
(2S,3S)-bis(diphenylphosphino)butane), but found only a
2.1:1 ratio of kinetic diastereomers (those with one hydride
trans to P and the other hydride trans to CO) at -25 °C. A
more extensive study by Kimmich et al. using [Ir(COD)-
(P-P*)]+ complexes found that at -80 °C the ratio of
diastereomers formed on oxidative addition was >50:1 for
the (R)-BINAP ((R)-BINAP ) (R)-2,2′-bis(diphenylphos-
phino)-1,1′-binaphthyl) derivative and nearly that large for
the (R,R)-Me-DUPHOS analogue ((R,R)-Me-DUPHOS )
(-)-1,2-bis((2R,5R)-2,5-dimethylphospholano)benzene)). Sub-
sequent conversions to thermodynamic isomers were also
followed. In contrast, Yamagata and co-workers observed
no diastereoselectivity in the reaction of H2 with Ir(Cl)(PPh3)-
(S-BINAP), obtaining only a 1:1 mixture of diastereomers.
Surprisingly, the largest diastereoselectivity for H2 oxidative
addition to an Ir(I) complex was obtained by Shin and Parkin
who examined a system containing a pair of chiral mono-
dentate phosphole ligands, P*, in a Vaska-type complex of
formula trans-IrCl(CO)(P*)2. The observed diastereoselec-
tivity for the meso isomer IrCl(CO)(R-P*)(S-P*) in which
the iridium is a “pseudoasymmetric center” was 60:1, with
the reaction proceeding under kinetic control as the initially
favored IrH2Cl(CO)(R-P*)(S-P*) isomer converts to an
equilibrium mixture favoring the other isomer.
oxidative addition, which occurs early along the reaction
coordinate for this exothermic reaction.
Catalytic asymmetric hydrogenation is an important class
of reactions that has been found to be particularly useful in
preparing biologically relevant chiral compounds such as
L-dopamine8 for treatment of Parkinson’s disease and as-
partame that contains L-phenylalanine as an artificial sweet-
ener.11 The hydrogenation catalysts used in the preparation
of both of these compounds are cationic Rh(I) systems
containing chiral bidentate di(phosphine) ligands. The mech-
anism of asymmetric hydrogenation using Rh(I) catalysts of
this type has been studied in detail by Halpern and Landis
who found that (a) reversible binding of the prochiral olefin
substrate is followed by irreversible reaction with H2, leading
to product and (b) the catalyst-olefin complex present in
minor amounts reacts with H2 much more rapidly than the
major diastereomer (by a factor of around 600).12-17 Hence,
from a kinetics standpoint, hydrogen activation is the
enantioselective step in the catalysis with ee’s for products
in excess of 95%.
These studies elegantly illustrate the importance of the
detection of minor species in equilibria if an accurate picture
of the catalytic pathway is to be assembled. Brown demon-
strated by NMR spectroscopy that the less stable enamide
isomer reacts substantially faster with H2 and was responsible
for the observed enantiomeric selectivity and, more recently
with Bargon, used p-H2 polarization methods to re-examine
this process by sensitizing the NMR experiment. They found
in this study that an agostic interaction exists in the metal
dihydride alkene intermediate.18
In this paper, we report the synthesis of two new chiral
Ir(I) complexes containing the (R)-BINAP ligand and their
reactions with H2. The study of the H2 oxidative addition
chemistry of these complexes is augmented by the use of
parahydrogen-induced polarization (PHIP) that can provide
greater sensitivity in analyzing reactions of this type.23-27
PHIP occurs when hydrogen enriched in the para spin state
(p-H2) adds pairwise to a compound in such a way that the
two 1H nuclei in the product are magnetically inequivalent.
Evidence of PHIP is seen by enhanced absorption and
emission lines in product 1H NMR spectra. PHIP is a
dynamic effect, and for it be observed, oxidative addition/
reductive elimination of H2 must occur fast relative to the
relaxation of the product hydride protons. The present study
For model Ir(I) complexes containing chiral di(phosphine)
ligands, H2 oxidative addition would be expected to generate
pairs of diastereomers since the cis dihydride product is chiral
(19) Kunin, A. J.; Farid, R.; Johnson, C. E.; Eisenberg, R. J. Am. Chem.
Soc. 1985, 107, 5315-5317.
(20) Kimmich, B. F. M.; Somsook, E.; Landis, C. R. J. Am. Chem. Soc.
1998, 120, 10115-10125.
(11) Fuganti, C.; Grasseli, P.; Malpezzi, L. J. Org. Chem. 1986, 51, 1126-
1128.
(21) Tani, J.; Nakajima, K.; Iseki, A.; Yamagata, T. Chem. Commun. 2001,
1630-1631.
(12) Landis, C. R.; Feldgus, S. Angew. Chem. Int. Ed. 2000, 39, 2863-
(22) Shin, J. H.; Parkin, G. J. Am. Chem. Soc. 2002, 124, 7652-7653.
(23) Duckett, S. B.; Sleigh, C. J. Prog. Nucl. Magn. Reson. Spectrosc. 1999,
23, 71-92.
2866.
(13) Landis, C. R.; Brauch, T. W. Inorg. Chim. Acta 1998, 270, 285-297.
(14) Chan, A. S. C.; Pluth, J. J.; Halpern, J. J. Am. Chem. Soc. 1980, 102,
5952-5954.
(15) Chua, P. S.; Roberts, N. K.; Bosnish, B.; Okrasinski, S. J.; Halpern,
J. J. Chem. Soc. Chem. Commun. 1981, 1278-1280.
(16) Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109, 1746-1754.
(17) Feldgus, S.; Landis, C. R. Organometallics 2001, 20, 2374-2386.
(18) Giernoth, R.; Heinrich, H.; Adams, N. J.; Deeth, R. J.; Bargon, J.;
Brown, J. M. J. Am. Chem. Soc. 2000, 122, 12381-12382.
(24) Eisenberg, R. Acc. Chem. Res. 1991, 24, 110-116.
(25) Eisenberg, R.; Eisenschmid, T. C.; Chinn, M. S.; Kirss, R. U. AdV.
Chem. Ser. 1992, 230, 47-74.
(26) Natterer, J.; Bargon, J. Prog. Nucl. Magn. Reson. Spectrosc. 1997,
31, 293-315.
(27) Bowers, C. R. In Encyclopedia of Nuclear Magnetic Resonance; Grant,
D. M., Harris, R. K., Eds.; John Wiley and Sons: Chichester, 2002;
Vol. 9: Advances in NMR, pp 750-770.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1197

Atesin et al.
Table 1. Crystallographic Data for IrI(CO)((R)-BINAP) (1) and
[Ir(CO)2((R)-BINAP)][SbF6] (2)
1
2
emp. formula
fw
C45H32P2OIIr
969.75
C46H32P2O2SbF6Ir
1106.61
T, K
100.0(1)
100.0(1)
λ, Å
0.71073
tetragonal
P41212
0.71073
monoclinic
P21
cryst syst
space group
Z
4
2
a, Å
b, Å
c, Å
12.0023(10)
12.0023(10)
26.132(4)
90
3764.4(7)
1.711
11.4905(15)
11.7732(15)
17.234(2)
102.564(2)
2275.6(5)
1.615
â, deg
V, Å3
F
calcd, mg/m3
µ, mm-1
4.486
3.644
abs correction
transm range
F(000)
SADABS
0.3523-0.8409
1880
SADABS
0.3620-0.8388
1072
2θ range, deg
1.87-30.99
-17 e h e17
-17 e k e 16
-37 e l e 37
59810
1.82-29.13˚
-15 e h e 15
-16 e k e 15
-23 e l e 23
33354
limiting indices
no. of reflns collected
no. of data/restraints/params
GOF
5921/25/250
1.010
12078/34/527
1.013
R1, wR2 (I > 2σ)
R1, wR2 (all data)
0.0353, 0.0598
0.0729, 0.0689
0.0343, 0.0796
0.0438, 0.0825
that the reaction be initiated at -70 °C under a CO
atmosphere and only slowly warmed to room temperature.
However, in the case of 1 with BINAP, which forms a
7-membered chelate ring on complexation, no evidence for
the bis chelate species was observed, even when BINAP was
present in excess as seen by 31P{1H} NMR spectroscopy.
The structure of 1 was confirmed by a single-crystal X-ray
diffraction study. Red crystals suitable for X-ray diffraction
were grown by slow diffusion of hexanes into a toluene
solution of 1. POV-RAY drawings of the molecular structure
of 1 are shown in Figure 1 along with selected bond lengths
and angles around the Ir(I) center. A complete tabulation of
metrical data for the structure is available in CIF format in
the Supporting Information. The molecule lies on a crystal-
lographic 2-fold axis (see Table 1), meaning that disorder
must exist in the structure (see the Experimental Section for
details of modeling the disorder). The essence of the disorder
lies in the carbonyl and iodide positions since these two
ligands have steric profiles that are not widely different from
a crystallographic standpoint. The BINAP ligand therefore
determines the crystal packing arrangement with a 50:50
disorder of iodide and CO, and because of the different bond
lengths involving these ligands with the metal center, the
iridium atom is also disordered.
Figure 1. POV-RAY images of 1 showing (a) the labeling of the atoms
around the Ir atom and (b) the view through the IrP2 plane showing its
twist relative to the I-Ir-C(O) plane of 25.85(34)°. Displacement ellipsoids
are drawn at the 50% level. Hydrogens have been removed for clarity.
Selected bond distances (Å) and angles (deg): Ir-P(1), 2.428(1); Ir-P(1#),
2.134(1); Ir-I, 2.643(1); Ir-C(1), 1.861(1); P(1)-Ir-P(1#), 91.79(5);
P(1#)-Ir-C(1), 92.0(3); P(1)-Ir-I, 90.32(4); C(1)-Ir-I, 91.1(3); P(1)-
Ir-C(1), 160.8(4); P(1#)-Ir-I, 164.08(7).
also analyzes the kinetic and thermodynamic differentiation
observed for the possible diastereomeric products.
Results and Discussion
Synthesis and Characterization of IrI(CO)((R)-BINAP)
(1). Addition of a toluene solution of (R)-BINAP to an
equimolar solution of [Bu4N][IrI2(CO)2] at room temperature
yields a mixture of the mono- and dicarbonyl complexes 1
and IrI(CO)2((R)-BINAP), respectively. On filtration of the
solution through a pad of alumina, the dicarbonyl species
readily eliminates one of its CO ligands to afford the
monocarbonyl compound 1 in 74% yield after workup.
Recrystallization from toluene and hexanes gives 1 as a dark
red powder. The 31P{1H} NMR spectrum of 1 is composed
of two doublets arising from two inequivalent phosphine
atoms at δ 19.12 and 16.65, with JPP of 29.6 Hz. The
spectrum is consistent with those obtained for other, previ-
ously reported IrX(CO)(P-P) complexes where X ) Cl, Br,
I and P-P ) a chelating diphosphine.9,19
Under the ambient temperature reaction conditions de-
scribed for the synthesis of 1, the analogous reaction with
other chelating di(phosphines) had previously led to forma-
tion of the corresponding bis chelate species [Ir(CO)x-
(PP)2]+where x ) 0, 1 as the major products. The successful
formation of IrX(CO)(P-P) analogues of 1 had necessitated
While satisfactory final agreement has been reached in
refinement of the structure, the disorder model renders the
metrical parameters within the coordination sphere of only
modest value for comparison purposes. For example, the
Ir-P bond length of the phosphine trans to CO (2.428(1)
Å), and the Ir-C(O) bond length (1.86(1) Å) are similar to
those found in other Ir(I) square planar complexes having
CO trans to phosphine (2.328-2.434 Å for Ir-P and 1.804-
1.940 Å for Ir-C(O)),28-31 while the Ir-P bond length of
(28) Cleary, B. P.; Eisenberg, R. Organometallics 1992, 11, 2335-2337.
1198 Inorganic Chemistry, Vol. 46, No. 4, 2007

DiastereoselectiWe OxidatiWe Addition of Dihydrogen
Scheme 2
lengths and angles around the Ir(I) center. (A complete
tabulation of metrical data is located in the CIF file deposited
as Supporting Information.) The crystal structure determi-
nation confirms that the geometry around the Ir(I) center is
essentially square planar with a minor tetrahedral distortion.
Specifically, the two trans P-Ir-C(O) bond angles average
172.0(2)° and the dihedral angle between the planes defined
by Ir(C(O))2 and IrP2 is 11.8°. The Ir-P bond lengths (2.357-
(1), 2.364(1) Å) and the Ir-C(O) bond lengths (1.916(5),
1.870(5) Å) are similar to those found in other Ir(I)
complexes having a CO ligand trans to a P atom (2.328-
2.434 Å for Ir-P and 1.804-1.940 Å for Ir-C(O)).28-31 The
difference between the two Ir-C(O) bond lengths, although
small, is greater than expected. The reason for this difference
is not immediately apparent. One reason inferred from the
crystal structure might be the closer proximity of fluorine
atom F1, found in the anion, to the carbon atom in the longer
Ir-C bond than to the other carbon atom (2.811(11) vs
2.995(11) Å), although it must be noted that the significance
or indeed the presence of such an interaction cannot be
determined due to the disorder present in the anion.
Figure 2. POV-RAY view of 2 showing the labeling of the atoms around
the Ir atom. Displacement ellipsoids are drawn at the 50% level. Hydrogens
have been removed for clarity. Selected bond distances (Å) and angles
(deg): Ir-P(1), 2.357(1); Ir-P(2), 2.364(1); Ir-C(1), 1.916(5); Ir-C(2),
1.870(5); P(1)-Ir-P(2), 92.02(5); P(1)-Ir-C(1), 90.02(18); P(2)-Ir-C(2),
88.7(2); C(1)-Ir-C(2), 90.4(3); P(1)-Ir-C(2), 172.0(2); P(2)-Ir-C(1),
172.0(2).
the phosphine trans to the iodide ligand (2.134(1) Å) is
somewhat shorter than that found for similar compounds
having phosphine trans to a halide ligand (2.171-2.218
Å).21,32-34
A notable feature of the molecular structure of 1 is a
significant tetrahedral distortion with a trans P-Ir-C(O)
bond angle of 160.8(4)°, a trans P-Ir-I bond angle of
164.08(7)°, and a dihedral angle of 25.85(3)° between the
IrP2 plane and the I-Ir-C(O) plane, as shown in Figure 1.
Synthesis and Characterization of [Ir(CO)2((R)-BI-
NAP)][SbF6] (2). Compound 2 is synthesized by the addition
of a CH2Cl2 solution of AgSbF6 to an equimolar solution of
1 that has been purged with CO. Workup followed by
crystallization from CH2Cl2 leads to the formation of a pale
orange powder of 2 in 97% yield. The 31P{1H} NMR
spectrum of 2 exhibits only one singlet at δ 12.11, indicating
equivalence of the two phosphine donors of the BINAP
ligand. The IR spectrum shows two CO stretches at 2084
and 2032 cm-1, consistent with a cationic structure containing
two CO ligands occupying cis coordination sites.
Reaction of [Ir(CO)2((R)-BINAP)][SbF6] with H2. The
presence of the chiral ligand (R)-BINAP in square planar
complex 2 means that in concerted H2 oxidative addition
two diastereomers will be produced. For a given enantiomer
of the BINAP ligand, the diastereomers differ by virtue of
the chirality at the metal center for the cis,cis-[IrH2(CO)2-
((R)-BINAP)]+product. In light of the C2 symmetry of
complex 2, the products of H2 addition from above and below
the metal plane are related by symmetry, and hence it is
possible to visualize formation of the two diastereomers by
simply considering the possible approaches of H2 to one face
of the complex only, as depicted in Scheme 2.
The structure of 2 was confirmed by a single-crystal X-ray
diffraction study. Dark yellow crystals suitable for X-ray
diffraction were grown by slow diffusion of diethyl ether
into a methylene chloride solution of 2. A POV-RAY
drawing of the molecular structure of 2 is shown in Figure
2 along with a figure caption that contains selected bond
1
The H2 oxidative addition reaction was followed by H
and 31P{1H} NMR spectroscopies. Figure 3 shows the
hydride region of 1H NMR spectra taken 45 s and 2 h after
the introduction of H2 into the reaction solution. The reaction
proceeds rapidly, as evidenced by an immediate color change
from the light orange of 2 to pale yellow. The initial H2
addition process proceeds to completion, and once the
reaction has commenced, no Ir(I) species can be observed
by NMR spectroscopy throughout the remaining course of
reaction. The kinetic diastereomer 3A predominates initially,
forming in a ratio to 3B of >9:1. The 1H NMR spectrum in
(29) Dahlenburg, L.; Von Deuten, K.; Kopf, J. J. Organomet. Chem. 1981,
216, 113-127.
(30) Brym, M.; Jones, C.; Waugh, M. Dalton Trans. 2003, 2889-2893.
(31) Schumann, H.; Cielusek, G.; Pickardt, J.; Bruncks, N. J. Organomet.
Chem. 1979, 172, 359-365.
(32) Oster, S. S.; Jones, W. D. Polyhedron 2004, 23, 2959-2965.
(33) Simons, R. S.; Panzner, M. J.; Tessier, C. A.; Youngs, W. A. J.
Organomet. Chem. 2003, 681, 1-4.
(34) Yamagata, T.; Iseki, A.; Tani, K. Chem. Letters 1997, 12, 1215-
1216.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1199

Atesin et al.
Figure 3. 1H NMR spectra (500 MHz) illustrating the change in the hydride region for the reaction of 2 with H2 in CD2Cl2 as a function of time.
Table 2. Summary of Simulated Rate Constants for the Reaction of 2
with H2 at 25 °Ca
rate constant
k, M-1 s-1
rate constant
k, s-1
k1
k2
0.92 (6)
k-1
k-2
6.0 (2) × 10-3
8.6 (5) × 10-2
1.53 (1) × 10-4
a Simulation parameters: delta time ) 50 s; run time ) 50 000 s; flux
tolerance ) 2 × 10-2; iterations/point ) 1; integral tolerance ) 1 × 10-6
.
All rate constants were allowed to vary independently. Errors are indicated
in parentheses.
to obtain best-fit rate constants for the overall process given
in eq 1.35,36 KINSIM was used to obtain initial fits for
concentration data of [3A] and [3B] versus time (see
Supporting Information for detailed data), after which
FITSIM was employed to refine the rate constants using a
least-squares fit method. The calculated rate constants for
the two-step isomerization process are given in Table 2, while
the calculated fits of concentration versus time are shown
by the solid lines in Figure 4. The fit to the experimental
data was excellent with an R2 value of 1.0000. The calculated
rate constants are consistent with the observed diastereose-
lectivity (k1/k2 ) 10.7:1) and the equilibrium constant (Keq
) (k2 × k-1)/(k1 × k-2) ) 3.67).
Figure 4. Change in concentration of 3A (9) and 3B(b) formed in the
reaction of 2 with H2 as a function of reaction time. The solid line reflects
the simulated trace assuming an exponential relaxation to equilibrium.
Figure 3 that was taken after 45 s shows that the two major
hydride resonances of 3A appear as a doublet of doublets at
δ -9.2 (Hb) due to almost equivalent couplings to two cis
phosphine donors (18 and 19 Hz), and a doublet of doublets
(Ha) at δ -10.62 arising from cis and trans phosphorus
couplings (16 and 112 Hz, respectively). Over a period of a
few hours, the ratio of 3A/3B decreases. The hydride
resonances for diastereomer 3B appear at δ -8.9 (Hd, JPH
of 16 and 17 Hz), and -10.6 (Hc, JPH of 12 and 114 Hz)
and exhibit similar couplings to those for 3A, in accordance
with the presence of identical coordination geometries. The
system relaxes to equilibrium by simple exponential behavior
and reaches a final diastereomeric ratio of 1:3.6 (Figure 4).
Extrapolation back to t ) 0 yields a high initial diaste-
reoselectivity for H2 oxidative addition of 9.9:1, which
3A yk1z 2 + H2 yk2z 3B
(1)
k-1
k-2
Reaction of IrI(CO)((R)-BINAP) with H2 and D2. The
oxidative addition of H2 to the iodo carbonyl complex 1 leads
to the formation of two diastereomeric pairs of dihydride
products corresponding to the kinetic and thermodynamic
isomers shown in Scheme 3. Since complex 1 lacks the C2
symmetry of dicarbonyl complex 2, the two faces of the
square plane to which H2 can approach are different and will
lead to opposite chirality at the Ir center when H2 is oriented
parallel to a specific axis of the complex. In the diastereo-
meric pair corresponding to the kinetic isomers (4A and 4B),
one hydride ligand is trans to CO and the second is trans to
a phosphine donor. In accord with what has been reported
previously for other IrX(CO)(P-P) systems, a greater than
95% stereoselectivity is seen for formation of the kinetic
diastereomers over the thermodynamic diastereomers.19 The
corresponds to a ∆∆G‡ of >1.36 kcal/mol for the two
298
different oxidative addition pathways leading to 3A and 3B.
Moreover, the observed equilibrium ratio of 1:3.6 corre-
sponds to a ∆G298 of -0.8 kcal/mol for the 3A h 3B
isomerization. The isomerization mechanism for intercon-
verting Ir(III) cis dihydride products from H2 oxidative
addition has previously been considered in terms of a simple
reductive elimination/oxidative re-addition sequence and a
dihydride transfer path at higher Ir and lower dihydrogen
concentrations.10 Simulations for the reductive elimination/
oxidative re-addition isomerization pathway were carried out
for the 2 + H2 system using the KINSIM/FITSIM programs
(35) Barshop, B. A.; Wrenn, R. F.; Frieden, C. Anal. Biochem. 1983, 130,
134-135.
(36) Zimmerle, C. T.; Frieden, C. Biochem. J. 1989, 258, 381-387.
1200 Inorganic Chemistry, Vol. 46, No. 4, 2007

DiastereoselectiWe OxidatiWe Addition of Dihydrogen
Figure 5. 1H NMR spectra (500 MHz) of the hydride region for the reaction of IrI(CO)((R)-BINAP) (1) with H2 in C6D6. (b, 4A; 9, 4B; 2, 5A; [, 5B).
Scheme 3
Scheme 4
Figure 6. Change in concentration of 4A (b), 4B (9), 5A (2), and
5B ([) formed in the reaction of 1 with H2 as a function of reaction
time. The solid line reflects the simulated trace based on the mechanism in
Scheme 4.
Table 3. Summary of Rate Constants Used to Simulate the Reaction of
1 with H2 at 25 °C
reaction is very rapid with an immediate color change from
red-orange to pale yellow once H2 is introduced into the
rate constant
k, M-1 s-1
rate constant
k, s-1
solution. Additionally, as in the oxidative addition of H2 to
2 to form 3A and 3B, a high degree of diastereoselectivity
is seen with an initial ratio of 4A/4B of 8.7:1 upon
extrapolation to t ) 0. However, after 20 min, 4B becomes
the predominant species (Figure 5) with an equilibrium ratio
of 4A/4B of 1:1.2.
The second pair of diastereomers corresponding to ther-
modynamic isomers 5A and 5B grow in over a period of
days. The upfield chemical shift in the hydride resonance
that shows cis coupling to both 31P nuclei is indicative of
the hydride trans to the iodide ligand.37 Additionally, within
the pair of thermodynamic diastereomers, there exists a
significant energetic differentiation, with the more stable
k1
k2
k3
k4
0.17
k-1
k-2
k-3
k-4
7.5 × 10-3
7.5 × 10-4
7.5 × 10-6
3.2 × 10-6
2.2 × 10-2
2.8 × 10-4
1.7 × 10-3
isomer being favored by a factor of 10:1. The degrees of
kinetic and thermodynamic differentiation seen within each
diastereomeric pair of the same IrH2I(CO)((R)-BINAP)
complex is impressive and shows that even for a molecule
as small as H2, there can exist a significant preference for
approach to one of the faces of 1 relative to the other. The
results with 1 + H2 contrast with what was seen with the
analogous chiraphos complex that showed a kinetic diaste-
reoselectivity of only 2.1:1 at -25 °C and an equilibrium
ratio of the thermodynamic diastereomers of 1.3:1.19
(37) Kaesz, H. D.; Saillant, R. B. Chem. ReV. 1972, 72, 231-281.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1201

Atesin et al.
Figure 7. Reaction of the dihydride isomers of 1 with para-enriched hydrogen in C6D6 (a) kinetic isomers at room temperature and (b) all four isomers at
85 °C.
Deuterium exchange with the kinetic and the thermody-
namic dihydride diastereomers of 1 was also studied. A
sample containing mainly the kinetic diastereomers 4A and
4B was prepared by shaking an NMR tube containing H2
and 1 in C6D6 for 30 min. At this stage the H2 in the NMR
tube was removed by one freeze-pump-thaw cycle, fol-
lowed by D2 addition. Another sample containing mainly
the thermodynamic diastereomers 5A and 5B was prepared
in an analogous fashion by keeping the H2 in the NMR tube
for 4 days before the introduction of D2 into the sample.
The hydride resonances of 4A and 4B rapidly decreased
in intensity and completely disappeared at room temperature
after 15 min and 3 h, respectively, indicating that, similar to
the oxidative addition of H2 to form the kinetic diastereomers,
the reductive elimination of H2 from them is also rapid. The
thermodynamic diastereomers on the other hand did not show
any appreciable exchange with D2 at room temperature and
required heating at 75 °C for a period of a couple of days
for complete exchange with D2.
that at ambient temperature, the isomerization of the kinetic
dihydride isomer of Ir(H)2Br(CO)(dppe) to its thermody-
namic counterpart involved both reductive elimination/
oxidative re-addition of H2 and a bimolecular pathway in
which the two hydrides of the kinetic isomer are transferred
to a molecule of IrBr(CO)(dppe) to give the thermodynamic
isomer while regenerating the Ir(I) species.10 The latter
pathway is dominant when less than 1 equiv of H2 relative
to IrBr(CO)(dppe) exists in the reaction system.
While both pathways were considered in modeling the
isomerization of diastereomers 4A, 4B, 5A, and 5B, only
the reductive elimination/oxidative addition sequence was
employed in the simulations. The bimolecular path was
excluded because 31P NMR measurements of the reaction
system revealed the absence of observable amounts of the
Ir(I) species 1 needed for the dihydride transfer mechanism
and the fact that its inclusion would cause the problem to
be underdetermined with up to 16 rate constants. Simulations
for the isomerization were therefore based solely on the
reductive elimination/oxidative addition mechanism shown
in Scheme 4 with eight rate constants.
As with the 2 + H2 reaction system, the mechanism of
isomerization for 4A, 4B, 5A, and 5B was probed by
simulating the experimental data of the 1 + H2 reaction using
the KINSIM program. Kunin et al. had previously shown
It was quickly determined, however, that whereas the
kinetic diastereomeric ratio could be reasonably determined
1202 Inorganic Chemistry, Vol. 46, No. 4, 2007

DiastereoselectiWe OxidatiWe Addition of Dihydrogen
by exponential extrapolation to t ) 0, the rate constant k1
could not be obtained with any degree of accuracy because
the reaction was much more rapid than the observed
concentration vs time data (standard deviations of k1 were
always larger than the corresponding rate constants). There-
fore, k1 was given an arbitrary value, large enough to
represent the fast initial reaction and small enough to fit the
experimental data, and the other rate constants were varied
manually to get the best fit to the data. The rate constants
for the isomerization process are tabulated in Table 3, while
both the experimental and calculated data are shown in Figure
6. These manually adjusted rate constants correspond to an
initial kinetic diastereoselectivity (k1/k2) of 7.7, and equilib-
rium ratios of 1.3 and 14 for the kinetic (4B/4A ) (k2 ×
k-1)/(k1 × k-2) and thermodynamic isomers (5B/5A ) (k4
× k-3)/(k3 × k-4), respectively. They compare favorably with
values of 8.7, 1.2, and 10 determined from the actual
measurements.
addition mechanism, the isomerization of the kinetic isomers
4A and 4B to the thermodynamic isomers 5A and 5B involve
a reductive elimination/oxidative addition and possibly other
bimolecular dihydride transfer steps. For both complexes 1
and 2, only the kinetically favored diastereomer shows any
polarization when parahydrogen was used.
Experimental Section
General Procedures and Materials. Unless otherwise stated,
all reactions and manipulations were performed in dry glassware
under a nitrogen atmosphere using either standard Schlenk tech-
niques or an inert-atmosphere glovebox. CH2Cl2, hexanes, and
diethyl ether were purified as described by Grubbs.38 Dichlo-
romethane-d2 and benzene-d6 were purchased from Cambridge
Isotope in ampules and used without further purification. (R)-BINAP
and AgSbF6 were purchased from Aldrich and used without any
further purification. Parahydrogen was prepared by cooling high-
purity hydrogen over FeCl3 adsorbed onto silica at 77 K.39 All NMR
spectra were recorded on a Bruker Avance 500 MHz spectrometer.
1H chemical shifts (δ in ppm) are referenced using chemical shifts
of residual solvent resonances. 31P chemical shifts (δ in ppm) are
relative to an external 85% solution of phosphoric acid in the
appropriate solvent. Mass spectrometric data were obtained on a
Hewlett-Packard Series 1100 MSD fitted with an atmospheric
pressure ionization chamber. Elemental analyses were performed
by Desert Analytics, Inc.
While Scheme 4 is sufficient to model most aspects of
the experimental data, it is evident from Figure 6 and the
differences between the observed and calculated kinetic
diastereoselectivity and thermodynamic isomer equilibrium
ratios that other steps, such as the binuclear isomerization
pathway, might also be playing a role in the conversion of
the kinetic isomers to the thermodynamic isomers.
IrI(CO)((R)-BINAP) (1). A 10 mL toluene solution of (R)-
BINAP (0.410 g, 1.32 mmol) was added to a 20 mL toluene solution
of [TBA][IrI2(CO)2] (0.490 g, 1.32 mmol). After the mixture was
stirred for 15 min, the color of the solution turned to dark orange
and [TBA]I precipitated. After filtration through a pad of alumina,
a dark red colored solution is obtained. Upon addition of hexanes,
1 is obtained as a reddish powder in a 74% yield. 1H NMR (C6D6,
ppm): δ 8.35-6.41 (overlapping naphthyl and aryl H); 31P{1H}
NMR (C6D6, ppm): δ 19.12 (d, JPP ) 29.6 Hz, 1 P), 16.65 (d, JPP
) 29.6 Hz, 1 P). Anal. Calcd for C45H32P2OIIr: C, 55.73; H, 3.33.
Found C, 55.67; H, 3.26.
[Ir(CO)2((R)-BINAP)][SbF6] (2). CO is bubbled through a 10
mL CH2Cl2 solution of 1 (0.337 g, 0.347 mmol). The color of the
solution turns to light orange after which a 5 mL CH2Cl2 solution
of AgSbF6 (0.120 g, 0.347 mmol) is added. AgI is removed by
filtration, and complex 2 is precipitated as a pale orange powder
Reaction of IrI(CO)((R)-BINAP) and [Ir(CO)2((R)-
BINAP)][SbF6] with Parahydrogen. When para-enriched
H2 is added to an NMR tube containing 1 dissolved in C6D6,
only 4A, the initially formed kinetic isomer, shows any
polarization at room temperature (Figure 7a). Upon heating
to 85 °C, 4B also starts to show polarization (Figure 7b).
When parahydrogen is introduced to a solution containing
the thermodynamic isomers 5A and 5B, no polarization is
observed even upon heating to 100 °C.
A similar situation arises when para-enriched dihydrogen
is added to a CD2Cl2 solution of 2. Only isomer 3A shows
any polarization, while isomer 3B shows no polarization even
upon heating. Owing to refluxing of the NMR solvent,
temperatures above 60 °C were not accessible with this
system. These PHIP results for the Ir BINAP complexes 1
and 2 are consistent with the observations of dihydrogen
addition discussed above and indicate that with H2 aligned
parallel to a specific axis of the complex, approach from
one side of the complex is favored over approach from the
other for oxidative addition and the corresponding reductive
elimination.
by the addition of diethyl ether, in a 97% yield. IR (KBr, cm-1
)
1
2084 (s, CO), 2032 (s, CO); H NMR (CD2Cl2, ppm): δ 7.92-
7.83 (m, 4 H), 7.77-7.67 (m, 10 H), 7.45 (t, 7.3 Hz, 2 H), 7.49-
7.42 (m, 2 H), 7.42-7.32 (m, 4 H), 7.3 (t, 7.5 Hz, 2 H), 7.04 (t,
7.2 Hz, 2 H), 6.91-6.82 (m, 4 H), 6.71 (d, 8.5 Hz, 2 H); 31P{1H}
NMR (CD2Cl2, ppm): δ 12.11 (s); MS (e/z) 871.3 ([Ir(CO)2((R)-
BINAP)]+). Anal. Calcd for C46H32P2O2SbF6Ir: C, 49.92; H, 2.91.
Found C, 50.16; H, 2.76.
Conclusions
General Procedure for the Reaction of 1 or 2 with H2 or Para-
Enriched Hydrogen. A resealable NMR tube was charged with 1
(0.005 g) and 0.6 mL of C6D6 (CD2Cl2 for 2). After the solution
was subjected to two freeze-pump-thaw cycles, the NMR tube
was back-filled with H2. The valve was closed, the solution was
allowed to thaw, and the NMR tube was then shaken vigorously to
promote solution-gas mixing. The reaction was then monitored
The oxidative additions of dihydrogen to the chiral Ir(I)
BINAP complexes 1 and 2 proceed rapidly at room tem-
perature and under kinetic control with a high degree of
diastereoselectivity, surpassing those of the previously
reported chiraphos analogue. For both 1 and 2, the kinetically
favored dihydride addition products are not the thermody-
namically favored isomers, and thermodynamic selectivity
is also observed, especially with complex 1. While the
isomerization of 3A to 3B can be modeled accurately
assuming only a simple reductive elimination/oxidative re-
1
periodically by H and 31P NMR spectroscopies.
(38) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518-1520.
(39) Millar, S. P.; Jang, M.; Lachicotte, R. J.; Eisenberg, R. Inorg. Chim.
Acta 1998, 270, 363-375.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1203

Atesin et al.
1
3A. H NMR (CD2Cl2, ppm): δ 8.20-6.65 (naphthyl and aryl
KR radiation (graphite monochromator). Final cell constants were
calculated from the xyz centroids of 3536 strong reflections from
the actual data collection after integration.
H’s overlapping with those of 3B), -9.23 (dd, JPP ) 18.3 Hz, 10.2
Hz, 1 H), -10.62 (dd, JPP ) 113.1 Hz, 14.9 Hz, 1H); 31P{1H}
NMR (CD2Cl2, ppm): δ 0.09 (d, JPP ) 15.2 Hz, 1 P), -2.70 (d,
JPP ) 17.6 Hz, 1 P).
Each structure was solved using SHELXS-9741 and refined using
SHELXL-97.41 Space groups P41212 (1) and P21(2) were deter-
mined on the basis of systematic absences and intensity statistics.
For each determination, a direct-methods solution was calculated
which provided most non-hydrogen atoms from the E-map. Full-
matrix least-squares/difference Fourier cycles were performed which
located the remaining non-hydrogen atoms. All non-hydrogen atoms
were refined with anisotropic displacement parameters. All hydro-
gen atoms were found from the difference map and refined with
relative isotropic displacement parameters. A summary of experi-
mental and refinement details can be found in Table 1.
1
3B. H NMR (CD2Cl2, ppm): δ 8.20-6.65 (naphthyl and aryl
H’s overlapping with those of 3A), -8.89 (dd, JPP ) 15.9 Hz, 11.1
Hz, 1 H), -10.50 (dd, JPP ) 113.1 Hz, 13.2 Hz, 1 H); 31P{1H}
NMR (CD2Cl2, ppm): δ 10.6 (d, JPP ) 18.4 Hz, 1 P), -2.70 (d,
JPP ) 15.3 Hz, 1 P).
4A. 1H NMR (C6D6, ppm): δ 8.55-6.30 (naphthyl and aryl H’s
overlapping with those of 4B, 5A, and 5B), δ -8.67 (dd, JPP
23.9 Hz, 14.2 Hz, 1 H), -10.83 (dd, JPP ) 149.6 Hz, 12.3 Hz, 1
H); 31P{1H} NMR (C6D6, ppm): δ -6.95 (d, JPP ) 7.4 Hz, 1 P),
-3.91 (d, JPP ) 7.4 Hz, 1 P).
4B. 1H NMR (C6D6, ppm): δ 8.55-6.30 (naphthyl and aryl H’s
)
Details for the Structure of 1. The molecule lies on a
crystallographic 2-fold axis; thus, one-half of the molecule is unique.
The iridium atom and the iodide and carbonyl ligands are disordered
over the 2-fold axis (50:50). Additionally, one of the phenyl rings
is modeled as disordered over two positions (50:50, because it is
coupled to the I/CO disorder).
Details for the Structure of 2. The cation and anion are well
separated, and all atoms lie on general positions. The anion is
modeled as disordered over two positions (58:42).
overlapping with those of 4A, 5A, and 5B), δ -8.51 (dd, JPP
)
18.2 Hz, 14.1 Hz, 1 H), -10.61 (dd, JPP ) 150.1 Hz, 11.2 Hz, 1
H); 31P{1H} NMR (C6D6, ppm): δ 16.22 (d, JPP ) 7.4 Hz, 1 P),
-3.51 (d, JPP ) 7.4 Hz, 1 P).
5A. 1H NMR (C6D6, ppm): δ 8.55-6.30 (naphthyl and aryl H’s
overlapping with those of 4A, 4B, and 5B), δ -9.88 (ddd, JPP
)
137.5 Hz, 18.5 Hz, JHH ) 4.8 Hz 1 H), -14.96 (ddd, JPP ) 23.6
Hz, 12.1 Hz, JHH ) 4.6 Hz 1 H); 31P{1H} NMR (C6D6, ppm): δ
-2.45 (d, JPP ) 18.75 Hz, 1 P), -5.65 (d, JPP ) 18.75 Hz, 1 P).
5B. 1H NMR (C6D6, ppm): δ 8.55-6.30 (naphthyl and aryl H’s
Acknowledgment. We gratefully acknowledge Prof. W.
D. Jones, Prof. H. A. Stern, and T. A. Atesin for assistance
with the kinetic simulations and Dr. D. J. Fox for helpful
discussions. We also thank the National Science Foundation
(Grant No. CHE-0092446) and the donors of the Petroleum
Research Fund for support of this work.
overlapping with those of 4A, 4B, and 5A), δ -9.92 (ddd, JPP
)
130.5 Hz, 14.6 Hz, JHH ) 3.9 Hz 1 H), -14.51 (td, JPP ) 13.3 Hz,
JHH ) 3.7 Hz 1 H); 31P{1H} NMR (C6D6, ppm): δ 11.72 (d, JPP
18.64 Hz, 1 P), -15.49 (d, JPP ) 18.84 Hz, 1 P).
)
Supporting Information Available: Plots of the extrapolation
of experimental data and tabulated experimental vs calculated data
for the reactions of 2 + H2 and 1 + H2, crystal data, atomic
coordinates, bond distances, bond angles, and anisotropic displace-
ment parameters for 1 and 2 in CIF format. This material is available
free of charge via the Internet at http://pubs.acs.org.
Crystal Structure Determinations. For each structure deter-
mination, a crystal was placed onto the tip of a 0.1 mm diameter
glass fiber and mounted on a Bruker SMART APEX II Platform
CCD diffractometer (Mo KR radiation, graphite monochromator)
for data collection at 100.0(1) K. A preliminary set of cell constants
and an orientation matrix were calculated from 158 (1) and 472
(2) reflections harvested from three sets of 20 frames, oriented such
that orthogonal wedges of reciprocal space were surveyed: four
major sections of frames were collected with 0.50° (1) or 0.30° (2)
steps in ω at four different φ settings and a detector position of
-28° in 2θ. The intensity data were corrected for absorption.40 The
data collection for each determination was carried out using Mo
IC061471A
(40) The SADABS absorption correction program is based on the method
of Blessing; see Blessing, R. H. Acta Crystallogr. Sect A 1995, 51,
33-38.
(41) SHELXTL: Structure Analysis Program, Version 5.04; Siemens
Industrial Automation, Inc.: Madison, WI, 1995.
1204 Inorganic Chemistry, Vol. 46, No. 4, 2007